Gastrointestinal stromal tumors (GISTs) are rare but deadly mesenchymal tumors characterized by expression of CD117 and oncogenic gain-of-function mutations in c-KIT and PDGFRa. Therapy for GIST patients has changed significantly with the use of imatinib mesylate (IM). Despite the initial success of the drug in the metastatic setting, particularly for those tumors with mutations in the juxtamembrane region of KIT that have a clinical benefit exceeding 80%, disease progression remains a perplexing problem necessitating multimodality management to truly cure this disease, as has been done in retroviral therapy. We have made several notable discoveries during the previous funding period, including demonstrating that "wild-type" GISTs (both adult and pediatric) that lack oncogenic mutations in c-KIT and PDGFRa appear to be driven by IGF-1R abnormalities. We have also shown as part of a phase II trial to evaluate the neoadjuvant use of IM that a gene signature that includes KRAB-ZNF 91 subfamily members can predict tumor shrinkage in response to IM and that gene expression modification within this signature by RNAi approaches can greatly enhance the activity of IM. The overall objectives going forward are to take the genes obtained through our synthetic lethal siRNA screens and gene profiling studies, use them to map the sensitization network for targeted therapeutics relevant to GIST and design meaningful combinations of siRNAs with drugs, or drug combinations that can be rapidly translated to the clinic. We have used RNAi approaches to identify candidates that selectively enhance killing by IM. Mapping the pattern of hits back to the network map revealed suggestive clusters of very closely interacting proteins, implying we have identified key survival nodes. The three Aims proposed will systematically develop our preliminary studies to identify productive targets of co-inhibition, with the ultimate goal of identifying new drug combinations that will greatly enhance the treatment of patients with GIST. Hence, Aim 1 will complete the initial hit validation process, establish which of the high-value set of candidate siRNAs sensitize a set of GIST cell lines to KIT and IGF-1R therapies and cluster these sensitizing genes into coordinated groups based on network modeling. In Aim 2 we will perform animal-based experiments to further test supersensitizing combinations of drug- and siRNA-based strategies. In Aim 3 we will test the novel therapeutic combination of imatinib and an anti-IGF1R targeted antibody in the treatment of patients with advanced GIST.